2949-1290 4 4 15 2025 1-102

UNK RUS 4-35 https://orcid.org/0000-0001-8679-7767 Ministry of Communication and Information Technology Muljono Wiryanta

Jakarta, Indonesia

Moestopo University Setiyawati Sri

Jakarta, Indonesia

Bandung Institute of Technology Setyanto Padmanabha Adyaksa

Bandung, Indonesia

The Moderating Role of Internet Accessibility on Consumer Online Shopping Intention in Indonesia The rapid advancement of digital technology has significantly shifted consumer behavior, leading to massive growth in Indonesia's e-commerce landscape. This study aims to develop and test a model that explains the formation of consumer Attitude and Online Shopping Intention by examining the influence of key antecedents: Convenience and Startup Credibility, while integrating the unique role of Internet Accessibility. The research is crucial in the context of developing nations like Indonesia, where varying internet infrastructure necessitates a deeper understanding of its impact. Using quantitative research methods, the data collected from internet users in major Indonesian economic regions (Jabodetabek, Joglosemar, and Gerbang Kertosusila) was analyzed using Partial Least Square (PLS). The findings confirmed that Convenience and Startup Credibility have a positive and significant influence on both Attitude and Online Shopping Intention. Furthermore, a positive Attitude is a significant predictor of Online Shopping Intention. Critically, the study found that Internet Accessibility acts as a significant moderator, strengthening the relationship between Convenience and Online Shopping Intention, Startup Credibility and Attitude, and Attitude and Online Shopping Intention. However, the moderating effect was not supported for the Convenience → Attitude and Credibility → Intention paths, suggesting that in certain high-penetration urban areas, accessibility may function more as a direct antecedent or is already perceived as adequate. This research provides a modified model, highlighting the critical importance of internet infrastructure (accessibility) as an uncontrollable but essential variable that fundamentally determines the success of e-commerce adoption in Indonesia. 10.57809/2025.4.4.15.1 330.47 online shopping intention convenience startup credibility attitude internet accessibility ecommerce https://technoeconomics.spbstu.ru/article/2025.15.1/ 1_muljono.pdf

UNK RUS 36-43 Peter the Great St.Petersburg Polytechnic University Klimentov Andrei

Saint Petersburg, Russia

Predicting claims in auto insurance using deep neural networks In the modern world, the insurance market is subject to significant changes, including under the influence of the use of digital technologies and the introduction of machine learning methods in insurance scoring. The object of the study is a data set with records of insurance policies. The study uses a deep nonlinear neural network to predict the occurrence of claim loss on auto insurance policies. Before using a multilayer neural network, data is preprocessed, and possible data leakage is eliminated. At the output of the neural network model, the resulting loss probability value is converted to a binary value. The model is evaluated using the ROC-AUC metric, with a graph of the ROC curve. The results show that the obtained model has predictive accuracy, but not high enough accuracy for industrial applications of the chosen model. The findings indicate the need for further research on ways to solve this problem using other machine learning methods. 10.57809/2025.4.4.15.2 330.47 machine learning neural networks insurance scoring prediction of insurance events auto insurance ROC-AUC classification scoring https://technoeconomics.spbstu.ru/article/2025.15.2/ 2_klimentov.pdf

UNK RUS 44-55 Peter the Great St.Petersburg Polytechnic University Kovalevskaya Daria

Saint Petersburg, Russia

N-1787-2013 56652307100 0000-0001-6251-7644 Imperial College London Svetunkov Sergey

London, The United Kingdom

A comparative analysis of machine learning methods with the application of the Kolmogorov-Gabor polynomial for forecasting sports event outcomes The article presents a comparative analysis of the effectiveness of machine learning methods for predicting the results of football matches, with a focus on the application of the elementary image of the Kolmogorov-Gabor polynomial. The relevance of the study is due to the need to choose models that are balanced in accuracy, interpretability, and computational complexity in conditions of high stochasticity of sports data. The scientific novelty lies in the adaptation of the elementary image of the Kolmogorov-Gabor polynomial (KGp) for sports analytics tasks and its complex comparison with a wide range of algorithms, from classical regression to gradient boosting. Based on historical data, models have been built and analyzed: an elementary image of a polynomial, linear regression with regularization, a random forest, gradient boosting, and a neural network. The results were evaluated by metrics MAE and accuracy of predicting the outcome. A model based on an elementary image of a polynomial Kolmogorov-Gabor showed competitive accuracy comparable to more complex ensemble methods, while maintaining advantages in computational efficiency and the potential interpretability of the structure of nonlinear dependencies. It was concluded that it is advisable to use this approach as an effective tool for building hybrid forecasting systems in sports analytics. 10.57809/2025.4.4.15.3 004.852 comparative analysis prediction of results football elementary image of the Kolmogorov-Gabor polynomial machine learning sports analytics gradient boosting random forest neural network regression analysis https://technoeconomics.spbstu.ru/article/2025.15.3/ 3_kovalevskaya_svetunkov.pdf

UNK RUS 56-69 Peter the Great St.Petersburg Polytechnic University Shmygol Tatyana

Saint Petersburg, Russia

Interrelation of business process maturity and spiral dynamics stages in enterprises The article explores the correlation between business process maturity levels and organizational development stages, as defined by the Spiral Dynamics framework. The research object is technology-intensive enterprises undergoing scaling and digital transformation. The methodological approach integrates maturity assessment tools, including process audits, structured interviews, and integrated maturity process index (MPI) calculation, with Spiral Dynamics diagnostics based on adapted questionnaires, hierarchy index analysis, and critical incident interviews. Results demonstrate a statistically significant correlation between process maturity and organizational value stages. Case analysis revealed that IT startups tended to shift from Orange to Green stages, while GMP-certified plants maintained Blue dominance despite advanced process maturity. The findings highlight cultural alignment as a decisive factor for successful digital transformation. Practical recommendations are proposed for phased standardization in early-stage organizations, cultural adaptation strategies in regulated enterprises, and conflict-resolution mechanisms in hybrid environments. 10.57809/2025.4.4.15.4 005.8 business process maturity spiral dynamics organizational development cultural alignment digital transformation process management IT startups regulated industries innovation hubs organizational culture https://technoeconomics.spbstu.ru/article/2025.15.4/ 4_shmygol.pdf

UNK RUS 70-80 Novosibirsk State Technical University Gumbo Kudakwashe

Novosibirsk, Russia

Predictive Maintenance in Helicopter Operations: Impact on Maintenance Cost, Safety, and Insurance This study explores the influence of predictive maintenance (PdM) on helicopter operations, focusing on its impact on manual inspections, operational costs, safety, and insurance. Using a dataset of maintenance-related aviation incidents, combined with statistical analysis in R, we uncover trends in incident frequency, injury severity, and fatality distribution over the past four decades. The results indicate that while overall incident rates have declined, the implementation of predictive maintenance correlates with measurable reductions in fatal and serious injuries, as well as operational costs and insurance liabilities. Our findings recommend broader adoption of PdM strategies, particularly in general aviation and helicopter fleets. 10.57809/2025.4.4.15.5 330.47 predictive maintenance helicopter operations aviation safety insurance maintenance cost manual inspection R analysis https://technoeconomics.spbstu.ru/article/2025.15.5/ 5_gumbo.pdf

UNK RUS 81-89 Sana'a University Al-sharabi Mohammed Abdullah Hayel

Sana'a, Yemen

Integrating Industry 5.0 Technologies into Enterprise Systems: A Sustainability Approach for Neste This case study explores the integration of Industry 5.0 technologies — Artificial Intelligence (AI), Internet of Things (IoT), and Blockchain—into the enterprise systems of Neste, a global leader in renewable fuels and circular economy solutions. The study examines how these technologies can enhance operational sustainability and support Environmental, Social, and Governance (ESG) goals by addressing specific gaps in the company’s digital infrastructure. Through a layered enterprise architecture analysis, the paper identifies opportunities for improving predictive capabilities, real-time monitoring, and ESG transparency. A three-phase roadmap is proposed to guide Neste’s transition toward a more human-centric and sustainable digital operating model. The study contributes to the literature by offering a practical, standards-aligned framework that supports long-term value creation and ESG compliance through technological innovation. 10.57809/2025.4.4.15.6 330.47 industry 5.0 sustainability environmental social governance artificial intelligence internet of things blockchain enterprise architecture digital transformation https://technoeconomics.spbstu.ru/article/2025.15.6/ 6_al_sharabi.pdf

UNK RUS 90-101 Peter the Great St.Petersburg Polytechnic University Ermochenko Semen

Saint Petersburg, Russia

Profit-risk optimization task for a hybrid warehouse configuration Object of study: hybrid warehouse architecture for e-commerce, integrating a physical warehouse and a virtual dropshipping channel. Methods: comparative analysis based on financial, operational, and risk-oriented indicators, supported by a mathematical framework incorporating supplier reliability. Results: the study reveals fundamental trade-offs between liquidity, risk, delivery speed, and costs. The hybrid model releases up to 40% of working capital but reduces profit by 25.3% at a supplier reliability of β = 0.95. Risk adjustment decreases expected profit by 11.25% compared to the nominal calculation. Conclusions: a verbal optimization problem is formulated to maximize profit under risk and delivery time constraints, providing a structured approach for managing hybrid systems instead of intuitive selection. 10.57809/2025.4.4.15.7 330.47 hybrid warehouse architecture inventory management e-commerce dropshipping supplier reliability risk management profit optimization supply chain logistics working capital order fulfillment multi-channel retail https://technoeconomics.spbstu.ru/article/2025.15.7/ 7_ermochenko.pdf